Posted 23 April 2020
I have a number of themes going on in my head today regarding coronavirus and it is unlikely that I could weave them together into a single piece. So I shall probably offer fragments during the day as I review and digest the evidence and then move on to the next theme.
Two points now concerning school closures.
Firstly an article published in the Lancet looking at the effectiveness of school closures. The abstract summarises that school closures contribute little to diminishing the transmission of SARS infections and contribute far less than other social distancing measures.
Data on COVID-19 is limited but their modelling is based on studies of other coronavirus outbreaks (MERS and SARS1). The key point that underpins the conclusion is the fundamental difference children play in the transmission of coronavirus compared to the influenza virus.
School closures and their effect on reducing influenza transmission is well studied and those studies confirm its effectiveness. However, children play a key role in influenza transmission: Infection rates of influenza are greater in children than adults and children contribute more to transmission than adults. Consequently school closures early enough in influenza outbreaks do reduce peak cases of those outbreaks.
Consistent data shows children have disproportionately low susceptibility (a term I used in yesterday’s post) to coronavirus compared to influenza. Children either do not become infected, or if they do, they are mostly asymptomatic. Although data is scarce, children seem to contribute far less to SARS transmission. Taiwan is cited as a country that has well-managed its coronavirus outbreak and chose not to close schools as part of that strategy.
A second piece of this particular jigsaw is a study of a nine-year old French girl who developed the COVID-19 infection. Through contact tracing, it was established that she came into contact with 172 other children and adults whilst still infectious. Not one of those contacts became infected.
That is a sample size n=1, but still useful information.